CareerPlan

Principal Ethernet Scale-Out Architect, AI Data Center

Santa Clara, California, United States of AmericaFull-time2026-09-17 → 2026-11-10

Core

Architect large-scale Ethernet-based AI/datacenter fabrics connecting servers, accelerators, racks, and multi-rack clusters.

Role type

Principal Hardware Scale-Out Architect (AI Data Center)

Builds

Next-generation scale-out network architectures for AI and cloud datacenter deployments

Domain

AI Infrastructure / High-Speed Networking / Data Center Hardware

Deliverable

production ML models

Required skills

Ethernet architecture, RoCE/RDMA, NIC/SuperNIC design, switch ASICs, scale-out interconnects, DMA engines, SerDes, PCIe/CXL interfaces, telemetry, RAS, security, virtualization

Preferred skills

PCIe/CXL/UCIe protocols, SONiC/SAI/OpenBMC/Linux networking, high-speed SerDes PHY design, signal integrity analysis

Technologies

Ethernet, RoCEv2, RDMA, InfiniBand, PCIe, CXL, UCIe, SONiC, SAI, OpenBMC, Linux

Responsibilities

Define next-generation scale-out network architectures; Develop end-to-end system architecture spanning accelerator cards to multi-rack clusters; Drive architecture for Ethernet fabrics, NICs, switch ASICs, and transport protocols; Analyze DMA engines and data flows to optimize bandwidth and latency; Collaborate with software and SoC teams to realize connectivity solutions; Provide technical leadership across physical, link, and transport layers

Seniority

Principal, hands-on IC with technical leadership